This application to extend our study "Meta-analysis for medicine: Applications and methods" encompasses two broad areas:(1) meta-analysis of interventions of medical and public health importance and (2) statistical investigations that aim to improve the quality of data collection, data analysis, and presentation in meta-analyses and in original research. We shall draw on datasets from ongoing meta-analyses for our methodologic work, and apply the results of this work in future meta-analyses. To fulfill these goals we intend to carry out seven specific aims: (1) Use our established procedures to compile data and carry out meta-analyses. Topics include: cardiovascular disease,oncology, dentistry, gastroenterology, rheumatology, epidemiological studies of cancer risk, and diagnostic imaging procedures. (2) Develop new methods and test proposed methods for meta-analysis to incorporate data from multiple outcomes reported in trials of a therapy. (3) Give effective techniques of graphical display for communicating results and analyzing data for meta-analysis and related statistical methods. (4) Refine systems for scoring quality of clinical trials and of diagnostic and epidemiologic studies. (5) Apply and develop methods of robust data analysis to address the problems and opportunities presented by heterogeneity. (6) Use the Surveillance, Epidemiology and End Results (SEER) database to quantify possible bias in estimates of efficacy of therapy -- using the example of breast cancer. (7) Examine approaches to updating meta-analyses as additional clinical trials are published by exploring the impact of study size and date of publication on clinical decisions that might be based on meta-analyses.